Search

We propose a constructive control design for stabilization of non-periodic trajectories of underactuated mechanical systems. An important example of such a system is an underactuated "dynamic walking" biped robot walking over rough terrain. The proposed technique is to compute a transverse linearization about the desired motion: a linear impulsive system which locally represents dynamics about a target trajectory. This system is then exponentially stabilized using a modified receding-horizon control design. The proposed method is experimentally verified using a compass-gait walker: a two-degree-of-freedom biped with hip actuation but pointed stilt-like feet. The technique is, however, very general and can be applied to higher degree-of-freedom robots over arbitrary terrain and other impulsive mechanical systems.

Hydrocephalus is related to a disturbed cerebrospinal fluid (CSF) system. For diagnosis, lumbar infusion test are performed to estimate outflow conductance, C (out), and pressure volume index, PVI, of the CSF system. Infusion patterns and analysis methods used in current clinical practice are not optimized. Minimizing the investigation time with sufficient accuracy is of major clinical relevance. The aim of this study was to propose and experimentally evaluate a new method, the oscillating pressure infusion (OPI). The non-linear model of the CSF system was transformed into a linear time invariant system. Using an oscillating pressure pattern and linear system identification methods, C (out) and PVI with confidence intervals, were estimated in real-time. Forty-two OPI and constant pressure infusion (CPI) investigations were performed on an experimental CSF system, designed with PVI = 25.5 ml and variable C (out). The ARX model robustly estimated C (out) (mean C (out,OPI) - C (out,CPI) = 0.08 μl/(s kPa), n = 42, P = 0.68). The Box-Jenkins model proved most reliable for PVI (23.7 ± 2.0 ml, n = 42). The OPI method, with its oscillating pressure pattern and new parameter estimation methods, efficiently estimated C (out) and PVI as well as their confidence intervals in real-time. The results from this experimental study show potential for the OPI method and supports further evaluation in a clinical setting.

Accurate estimates of the outflow resistance of the human cerebrospinal fluid system are important for the diagnosis of a medical condition known as hydrocephalus. In this paper we design a nonlinear observer which provides on-line estimates of the outflow resistance, to the best of our knowledge the first method to do so. The output of the observer is proven to globally converge to an unbiased estimate. Its performance is experimentally verified using the same apparatus used to perform actual patient diagnoses and a specially-designed physical model of the human cerebrospinal fluid system.

The problem is to create a hybrid periodic motion, reminiscent of walking, for amodel of an underactuated biped robot. We show how to construct a transverse linearization analytically and how to use it for stability analysis and for design of an exponentially orbitally stabilizing controller. In doing so, we extend a technique recently developed for continuous-time controlled mechanical systems with degree of underactuation one. All derivations are shown on an example of a three-link walking robot, modeled as a system with impulse effects.

This paper provides an introduction to several problems and techniques related to controlling periodic motions of dynamical systems. In particular, we consider planning periodic motions and designing feedback controllers for orbital stabilization. We review classical and recent design methods based on the Poincaré first-return map and the transverse linearization. We begin with general nonlinear systems and then specialize to a class of underactuated mechanical systems for which a particularly rich structure allows many of the problems to be solved analytically.

This paper introduces a method for analytical planning of feasible hybrid periodic trajectories in non-feedback-linearizable impulsive mechanical systems with control inputs. For a planned motion, a procedure for computation of a transverse linearization for a class of hybrid controlled mechanical systems with underactuation one is presented. The resulting linear comparison system can be used for stability analysis and for design of orbitally stabilizing controllers.

This paper presents a vision-based wheeled-robot navigation technique, termed circular navigation guidance (CNG), for the interception of a moving target from a specific angle. The guidance law is not split into path-planning and path-following stages, but is continuously updated based on immediately available information, making it useful even against a manoeuvring target, or where only incomplete information is available to the robot. This is achieved by converting a finite-time navigation problem into one of maintaining a certain geometric condition at all times. Detailed theoretical analysis provides a closed-form solution for the shape of the path the robot takes against a moving target. Experimental results are presented which show that CNG is robust and accurate under real-world conditions against a manoeuvring target.

A new precision guidance law is presented for three-dimensional intercepts against a moving target In contrast to previously published guidance laws, it does not require knowledge of the range to the target. This makes it appropriate for use on platforms which have an imaging device, such as a video camera, as a primary sensor. We prove that with idealized dynamic model, the guidance law results in zero miss distance, and a formula is given for impact angle error which tends to zero as does target speed, making this method particularly suitable against slow moving targets. Computer simulations are used to test the law with a more realistic model, with a video camera and optical-flow algorithm providing target information. It is shown to perform well compared with another law from the literature, despite requiring less information.

Recently, a new technique for generating periodic motions in mechanical systems which have less actuators than degrees of freedom has been proposed. A motivating example for studying such motions is a dynamically stabilized walking robot, where the target trajectory is periodic, and one of the joints - the ankle joint - is unactuated, or weakly actuated. In this paper, the technique is implemented on the Furuta pendulum, an experimental testbed that is simpler than a walking robot but retains many of the key challenges - it is underactuated, open-loop unstable, and practical problems such as friction and velocity estimation must be overcome. We present a detailed description of the practical implementation of the controller. The experiments show that the technique is sufficiently robust to be useful in practice.

Idiopathic normal pressure hydrocephalus (INPH) patients have a disturbance in the dynamics of the cerebrospinal fluid (CSF) system. The outflow conductance, C, of the CSF system has been suggested to be prognostic for positive outcome after treatment with a CSF shunt. All current methods for estimation of C have drawbacks; these include lack of information on the accuracy and relatively long investigation times. Thus, there is a need for improved methods. To accomplish this, the theoretical framework for a new adaptive observer (OBS) was developed which provides real-time estimation of C. The aim of this study was to evaluate the OBS method and to compare it with the constant pressure infusion (CPI) method. The OBS method was applied to data from infusion investigations performed with the CPI method. These consisted of repeated measurements on an experimental set-up and 30 patients with suspected INPH. There was no significant difference in C between the CPI and the OBS method for the experimental set-up. For the patients there was a significant difference, −0.84 ± 1.25 µl (s kPa)−1, mean ± SD (paired sample t-test, p < 0.05). However, such a difference is within clinically acceptable limits. This encourages further development of this new real-time approach for estimation of the outflow conductance.

We present a robot navigation technique, termed circular navigation guidance (CNG), which solves the challenging problem of intercepting a moving target from a specific angle, using only data from a video camera. The guidance law is not split into path-planning and path-following stages, but is continuously updated based on immediately available information, making it practical even against a manoeuvring target, or where only incomplete information is available to the robot. This is achieved by converting a finite-time navigation problem into one of maintaining a certain geometric condition at all tunes. Theoretical and experimental results are presented, which support CNG as a practically useful tool for robot navigation.

We present a new control law for the problem of docking a wheeled robot to a target at a certain location with a desired heading. Recent research into insect navigation has inspired a solution which uses only one video camera. The control law is of the "behavioral" type in that all control actions are based on immediate visual information. Docking success under certain conditions is proved mathematically and simulation studies show the control law to be robust to camera intrinsic parameter errors. Experiments were performed for verification of the control law.